SIMULTANEOUS DETERMINATION OF BENZOIC ACID AND SORBIC
ACID IN FOOD PRODUCTS BY CAPILLARY ELECTROPHORESIS
Nevin Öztekin
Istanbul Technical University, Department of Chemistry, Maslak, 34469 Istanbul, Turkey Submitted: 12.08.2017 Accepted: 12.12.2017 Published online: 13.03.2018 Correspondence: Nevin ÖZTEKİN E-mail: noztekin@itu.edu.tr ©Copyright 2018 by ScientificWebJournals Available online at www.scientificwebjournals.com ABSTRACT
In this work, a rapid and easy capillary electrophoretic method for simultaneous determination of benzoic acid and sorbic acid in food samples using direct UV detection was reported. The separation of acids was achieved in fused silica capillary at 28 kV. 20 mM borate buffer at pH 9.3 was used as running buffer; cinnamic acid was chosen as internal standard for quantification. Under the optimal separation conditions benzoic acid and sorbic acid were detected in 3.1 min. The linear ranges were between 0.005-0.4 mM. The correlation coefficients for each calibration curve were calculated as 0.999. The reproducibility of peak area and migration time for each acid were less than 3% (R.S.D.). The limits of detection were found 0.405 g/mL for benzoic acid and 0.415 g/mL for sorbic acid. The limits of quantification were 1.35 μg/mL and 1.38 μg/mL respectively.
Keywords: Capillary electrophoresis, Benzoic acid, Sorbic acid, Foods
Cite this article as:
Öztekin, N. (2018). Simultaneous Determination of Benzoic Acid and Sorbic Acid in Food Products by Capillary Electrophoresis. Food and Health, 4(3), 176-182. DOI: 10.3153/FH18018
Introduction
Benzoic acid (E210) and sorbic acid (E200) are widely used as food additives to delay the microbiological contamination due to action of bacteria, yeasts and moulds in various foods. They are supposed as usually reliable preservatives. If food products contain high concentrations of these acids, they may damage to health of consumers due to causing allergic reactions. Like the other countries, the use of these additives have been restricted according to Turkish Food Codex reg-ulations of Agriculture Ministry of Turkey. In order to pro-tect human health, determination of maximum permitted amounts of these additives in food products is very im-portant. Consequently, easy and reliable analysis methods for detection of these additives in foodstuffs is required for food safety. Several analytical methods are available for the determination of additives in foods like UV spectrometry (Marsili, et al., 2003; Naseri, et al. 2017) and GC (Ochiai et al., 2002; Abedi et al. 2014). Both of these methods require comprehensive sample preparation and relatively long anal-ysis time. HPLC is most commonly used method for the de-termination of these additives in foods and beverages (Mota et al., 2003; Pylypiw Jr & Grether, 2000; Tfouni & Toledo, 2002; Zor et al. 2016; Petanovska-Ilievska et al. 2017). HPLC generally needs the consumption of large amounts of organic solvents during the analysis and extensive sample preparation step, including solid phase extraction.
Capillary electrophoresis (CE) is an analytical method that offer high separation efficiency and is ensuring a fast sepa-ration for the assay of several species. CE method has some advantages such as fast analysis time, low or no organic sol-vent consumption and little injection volume. A matrix ef-fect in sample can be eliminated with short and simple sam-ple pretreatment. The capillary between runs is quickly washed with running buffer. Therefore it is a convenient analysis technique for using in the analysis of foods con-tained complex matrices.
The inhibitory action of weak acids is classically believed to be due to the undissociated compounds. Because, the un-charged, undissociated state of the acid is freely passed through the plasma membrane and is thus ensured to enter
the cell (Brul & Coote, 1999). pKa values of benzoic and
sorbic acid are smaller than pH 7 (4.19 and 4.76 respec-tively). The activity of them in foods that has value of low pH increase. These acids are used as a substitute for each other. Nevertheless, we observed the mixtures of benzoic and sorbic acid are used in some food products in Turkey. While several of the CE reports demonstrated the
possibili-1996; Han et al., 2008; Hsu et al., 2014; Sun et al., 2014; Li et al., 2015; Aung & Pyell 2016), only a few of them re-ported benzoic and sorbic acids together in real samples. This study presents an application for the quantification of benzoic and sorbic acid in the food samples in the Turkish markets by using CE method. Before the injection pro-cesses, any pretreatment procedure was not carried out for the food samples.
Materials and Methods
Instrumentation
For separations and determinations, an Agilent CE system (Waldbronn, Germany) equipped with a diode array detec-tor was used. Agilent Chem Station software performed all system control and data processing. In separations, 50m I.D. fused-silica capillary (Polymicro Technologies, Phoe-nix, AZ, USA) with 46 cm total length and 38 cm effective length was used. The capillary temperature was set at 25ºC. The separations were performed at voltage of 28 kV. The injection of samples was made at pressure 50 mbar for 4s. Before injections at the beginning of every day the capillary was conditioned with 0.1 M NaOH solution for 15 min, de-ionized water for 5 min and running buffer for 10 min. be-tween runs 1 min of flushing with running buffer was per-formed. A Metrohm 654 Digital pH Meter was used for pH measurements. All solutions were prepared with deionized water obtained from an Elgacan C114 filtration system. Chemicals
Benzoic acid, sorbic acid, and cinnamic acid were pur-chased from Fluka (Fluka AG, Buchs, Switzerland). Sodium borate was obtained from Merck (Darmstadt, Germany). Other chemicals used were of analytical reagent grade. The food samples were from local market.
Standards
Stock solutions of benzoic acid, sorbic acid, and cinnamic acid were prepared by dissolving in water, and stored at 4ºC. Samples
After solid samples were homogenized, 1-2 g samples were weighed in a beaker. 15 mL of deionized water was added to samples and stirred during 30 min. The volume of sample was diluted to 20 mL. The solutions were centrifuged a 7000 rpm for 5 min and filtered through a 0.45 μm syringe filter. Internal standard solution (cinnamic acid) was added to
fil-Results and Discussion
In CE analysis the electroosmotic flow (EOF) plays an im-portant role. The magnitude of the EOF can influence the resolution and separation efficiency. As we all know, the EOF can be determined by net surface charge of the inner surface of capillary. At higher pH (usually alkaline), faster EOF are obtained due to the full ionization of silanol groups. The concentration and pH of buffer are the two important parameters for adjusting the EOF and the electrophoretic mobility. Therefore the separation buffer affects the migra-tion time and the resolumigra-tion between analytes. The ioniza-tion degree of an organic acid also depends on choice of the buffer. At the same time the pH of the buffer influences ion-ization degree and separation of organic acids. Simultane-ous migrations of benzoic acid, sorbic acid, and cinnamic acid were tested in the three different separation buffer namely; borate, phosphate and Tris in the pH range of 7 and
9.5. Their pKa values are 4.19, 4.76 and 4.27, respectively
and in this pH range all acids are in higher ionization de-grees. Thus they can be separated as anions in alkaline me-dia.
When borate buffer was used as the separation media, peak heights and peak areas were observed as better than the peaks in the other two buffers, and also migration times were shorter. Therefore borate was selected as the running buffer. Dilute buffer solutions are desirable to shorten mi-gration time. However, it is important that buffer solutions have sufficient buffer capacity to maintain a fixed pH. Inad-equate buffer capacity may result in the poor resolution. The effect of borate concentration was studied within the range 10-50 mM. At higher borate concentrations (>50 mM) longer migration time and broader peak were obtained. Op-timum peak symmetry and peak area were achieved with 20 mM borate.
When the running voltages is increased the efficiency of the separation improves and the migration time shortens. In the higher running voltage, the efficiency of the separation de-crease due to the Joule heating. On the other hand, the re-producibility and the resolutions between acids becomes worse if the running voltage decreases. In this study a volt-age of 28 kV was selected as the best separation voltvolt-age with acceptable compromise between adequate resolution and short migration time.
In CE the high amounts of matrix components in the samples affect the peak shapes of analytes has generally low concentrations. In order to check the matrix effect on the peak shapes, the real samples was injected to the borate buffer with increasing concentrations. It was not observed a distortion in the peak shapes in the small borate concen-trations. Since the matrix effect was not observed in the highly diluted real samples, all separations were performed in the 20 mM borate buffer at pH 9.3, considering the fast separation times.
Figure 1 shows the absorption spectra of the analytes and internal standard in the same concentration level. In Figure 2 the peak areas of benzoic acid, sorbic acid, and cinnamic acid were given at the different wavelengths varying from 210 to 270 nm. According to peak areas 220 nm can be se-lected for the simultaneous screening of benzoic acid, sorbic acid, and cinnamic acid peaks. In this study because of us-ing a diode array detector for the detection, benzoic acid and sobic acid was dedected different wavelengths respectivly 220 nm and 260 nm.
The electropherogram of standard mixture of acids at opti-mum experimental conditions are shown in Figure 3a. The samples were injected directly to the borate buffer and all acids were detected in about 3.0 min.
Figure 2. The effect of the wavelength on the peak areas of benzoic acid (), sorbic acid () and cinnamic acid ().
Running buffer, 20 mM Borate, pH 9.3; injection, 4 s at 50 mbar; run voltage, 28 kV
Method Validation Linearity
The calibration curves were obtained between the peak area
ratios (Aanalyte/Ainternal standard) and the concentrations of the
ac-ids. The calibration curves for benzoic acid and sorbic acid showed linear ranges between 0.005 and 0.4 mM. Linear regression equations for benzoic acid and sorbic acid were
obtained as y = 8.140 10-1 x – 9.558 10-4 and y = 6.474 10-1
x – 3.459 10-3 respectively. The correlation coefficients
val-ues were greater than 0.999. Limit of detection
Limits of detection (LOD) were calculated as three times the baseline noise (S/N=3). The LOD value was 0.405 μg/mL for benzoic acid and 0.415 μg/mL for sorbic acid. Limits of quantification (LOQ) (S/N=10) was 1.35 μg/mL and 1.38 μg/mL respectively.
Reproducibility
The reproducibility of the CE method was determined by analyzing two real samples, Ketchup and Chestnut candy. % RSD values of migration times and peak areas were calcu-lated for intra-day and inter-day. Intra-day and inter-day precision were calculated seven successive injections in a day and three injections in different days, respectively. The results are given in Table 1.
Table 1. Reproducibilities of the method for two sample. (I) Ketchup, (II) Chestnut sweet
Intraday RSD% Inter day RSD% mg anion/kg sample
RSD %
Migration time Peak area Migration time Peak area Inter day Intra day
Sample 1 Benzoic 0.213 0.375 0.645 1.234 0.364 1.376 Sorbic 0.194 0.877 0.724 1.582 0.956 2.753 Sample 2 Benzoic 0.0741 0.638 0.445 2.398 0.744 1.942 Sorbic 0.0697 0.559 0.421 2.179 0.601 1.894
Table 2. Recovery of the method for two samples. (I) Ketchup, (II) Chestnut sweet
Added amount Sample (I) Recovery (%) Sample (II) Recovery (%)
Benzoic acid Sorbic acid Benzoic acid Sorbic acid
0.03 mM 97 98 104 101
0.06 mM 98 103 101 102
0.10 mM 103 100 99 101
Table 3. Results for the determination of benzoic acid and sorbic acid in various food products
Sample Benzoic acid (mg/kg or mg/L) Sorbic acid(mg/kg or mg/L)
Lemon sauce 4.36 ±0.12 235.5 ±0.9 Pickle 126.9 ±1.1 127.3 ±0.9 Fruit juice 234.9 ±0.9 178.5 ±0.7 Ketchup 463.2 ±1.0 75.3 ±1.8 Chestnut sweet 200.5 ±1.9 251.5 ±1.9 Olive paste 13.1 ±1.7 600.7 ±1.5 Pepper paste 495.5 ±1.8 10.5 ±0.4 Pepper sauce 182.8 ±1.6 188.5 ±1.2 Cake n.d. 600.7 ±2.0 n.d.: not detected
Recovery
The accuracy of method was evaluated with two methods: (1) The peak purity was analyzed with diode array detector and (2) the known amounts of acid standards were added to food samples. Table 2 shows the recovery results for three concentration levels and for two food samples.
Application to Food Samples
The benzoic acid and sorbic acid in nine food products were quantified by using the CE method. Figures 3b and 3c illus-trate the electropherograms of two food samples. In opti-mized conditions acids were detected within about 3.0 min. Amounts of preservatives determined in each sample were listed in Table 3. As seen from the electropherograms, there are unknown peaks in this region, but peaks of acids and internal standard are seen clearly. All food samples contain sorbic acid. In only one food sample benzoic acid was not detected. The amount of determined benzoic acid in food samples was between 4.36 and 495.5 mg/kg or mg/L. The content of sorbic acid in food samples ranged from 10.5 to 600.7 mg/kg or mg/L. The obtained data indicate that the preservatives content in foods were under permitted maxi-mum levels.
Conclusion
In this work a simple, easy and reliable CE method was de-scribed for the simultaneous determination of benzoic and sorbic acid in various food products. The linearity of cali-bration curves was very good (r>0.999) and the reproduci-bility of method were also satisfactory (RSD < 3%). Method contain both short sample preparation time and short analy-sis time (~3.0 min). The matrix of samples did not interfere the analysis of analytes. In addition it contain low electrolyte and sample consumption. The silica capillaries are cheaper than chromatographic columns, easily washed between runs. Therefore the proposed CE method is used both as al-ternative method to HPLC for the analysis of additives in food products and routine analysis of various food for sam-ples benzoic and sorbic acid.
References
Abedi, A.S., Mohammadi, A., Azadniya, E., Mortazavian, A.M., Khaksar, R. (2014). Simultaneous determination of sorbic and benzoic acids in milk products using an optimised microextraction technique followed by gas chromatography. Food Additives and Contaminants Part A, 31, 21-28
Aung, H.P., Pyell, U. (2016). In-capillary derivatization with o-phthalaldehyde in the presence of 3-mercapto-propionic acid for the simultaneous determination of monosodium glutamate, benzoic acid, and sorbic acid in food samples via capillary electrophoresis with ul-traviolet detection. Journal of Chromatography A, 1449, 156-165
Boyce, M.C. (1999). Simultaneous determination of antiox-idants, preservatives and sweeteners permitted as addi-tives in food by mixed micellar electrokinetic chroma-tography. Journal of Chromatography A, 847, 369-375.
Brul, S., Coote, P. (1999). Preservative agents in foods - Mode of action and microbial resistance mechanisms. International Journal of Food Microbiology, 50, 1-17. Frazier, R.A., Inns, E.L., Dossi, N., Ames, J.M., Nursten,
H.E. (2000). Development of a capillary electrophore-sis method for the simultaneous analyelectrophore-sis of artificial sweeteners, preservatives and colours in soft drinks. Journal of Chromatography A, 876, 213–220.
Han, F., He, Y.Z., Li, L., Fu, G.N., Xie, H.Y., Gan, W.E. (2008). Determination of benzoic acid and sorbic acid in food products using electrokinetic flow analysis–ion pair solid phase extraction–capillary zone electropho-resis. Analytica Chimica Acta, 618, 79-85.
Hsu, S.H., Hu, C.C., Chiu, T.C. (2014). Online dynamic pH junction–sweeping for the determination of benzoic and sorbic acids in food products by capillary electro-phoresis. Analytical and Bioanalytical Chemistry, 406, 635-641.
Kuo, K.L., Hsieh, Y.Z. (1997). Determination of preserva-tives in food products by cyclodextrin modified capil-lary electrophoresis with multiwavelength detection. Journal of Chromatography A, 768, 334-341.
Li, B., Li, Y.K., Wang, X.Y., Wang, F., Wang, X., Wang, Y.F., Meng, X.J. (2015). Simultaneous separation and determination of organic acids in blueberry juices by capillary electrophoresis- electrospray ionization mass spectrometry. Journal of Food Science and Technol-ogy, 52, 5228-5235.
chromatography. Journal of Chromatographic Sci-ence, 38, 345-352.
Marsili, N., Sobrero, M.S., Goicoechea, H.C. (2003). Spec-trophotometric determination of sorbic and benzoic ac-ids in fruit juices by a net analyte signal-based method with selection of the wavelength range to avoid non-modelled interferences. Analytical and Bioanalytical Chemistry, 376, 126-133.
Mota, F.J.M., Ferreira, I.M.P.L.V., Cunha, S.C., Beatriz, M., Oliveira, P.P. (2003). Optimization of extraction procedures for analysis of benzoic and sorbic acids in foodstuffs. Food Chemistry, 82, 469-473.
Naseri, A., Farahmand, F., Sheykhizadeh S., Tabibiazar M., (2017). Coupling of air‑assisted liquid–liquid microex-traction method with partial least squares for simulta-neous spectrophotometric determination of some pre-servatives. Journal of the Iranian Chemical Society, 14, 643-653.
Ochia, N., Sasamoto, K., Takino, M., ·Yamashita, S., Dai-shima, S., Heiden, A.C., Hoffmann, A. (2002). Simul-taneous determination of preservatives in beverages, vinegar, aqueous sauces, and quasi-drug drinks by stir-bar sorptive extraction (SBSE) and thermal desorption GC–MS. Analytical Bioanaytical Chemistry, 373, 56-63
Pant, I., Trenerry, C. (1995). The determination of sorbic acid and benzoic acid in a variety of beverages and foods by micellar electrokinetic capillary chromatog-raphy. Food Chemistry, 53 219-226.
Petanovska-Ilievska, B., Velkoska-Markovska, L., Janku-lovska, M.S. (2017). Development of reverse-phase high-performance liquid chromatography method for simultaneous determination of sodium benzoate and potassium sorbate in beverages. Acta Chromatograph-ica, 29, 345-358.
Pylypiw, Jr. H.M., Grether, M.T. (2000). Rapid high-perfor-mance liquid chromatography method for the analysis of sodium benzoate and potassium sorbate in foods. Journal of Chromatography A, 883, 299-304
Sun, B., Qi, L., Wang, M. (2014). Determination of preserv-atives in soft drinks by capillary electrophoresis with ionic liquids as the electrolyte additives. Journal of Separation Science, 37, 2248-2252.
Tfouni, S.A.V., Toledo, M.C.F. (2002). Determination of benzoic and sorbic acids in Brazilian food. Food Con-trol, 13, 117-123.
Waldron, K.C., Li, J. (1996). Investigation of a pulsed-laser thermo-optical absorbance detector for the determina-tion of food preservatives separated by capillary elec-trophoresis. Journal of Chromatography B, 683, 47-54.
Zor, Ş.D., Aşçı, B., Dönmez, Ö.A., Küçükkaraca, D.Y.
(2016). Simultaneous determination of potassium sorbate, sodium benzoate, quinoline yellow and sunset yellow in lemonades and lemon sauces by HPLC using experimental design. Journal of Chromatographic Sci-ence, 54, 952-957.
